Exciton trapping in dihalonaphthalene crystals

Journal of Photochemistry,

12 (1980) 1610 Elsevier Sequoia S.A., Lausanne -Printed

166 in Switzerland

161

Short Communication

Exciton trapping in dihalonaphthalene crystals

KAREN

J. LATAS,

Department

J. R. SIMONS

of Chemistry,

(Received March 8,1979;

and A.M.

NISHIMT_lRA

Wichita State University,

Wichita,

Kan.

67208

(U.S.A.)

in revised form June 6,1979)

Recent interest in the optical spectroscopy of dihalonaphthalenes has arisen because crystals of these molecules form excellent models of onedimensional linear excitons, particularly crystals of 1,4-dibromonaphthalene [ 1 - 71. In this communication we present a preliminary study of the phosphorescent triplet states in neat and chemicaIIy mixed dihalonaphthalenes of 1,4-dichloronaphthalene (DCN), l-bromo-4chloronaphthalene (BCN) and 1,4_dibromonaphthalene (DBN) . Optically detected magnetic resonance (ODMR) techniques were used to identify localized states in single crystals of the dihalonaphthalenes. Initially the triplet states of the dihalonaphthalenes were characterized by the total phosphorescence rate constants, by the triplet state zero field splittings (zfs) and by the vibronic progression in the phosphorescence spectra. In the dihalonaphthalenes these parameters are characteristic of a particular molecule. In the mixed crystal studies the identity of the individual emission peaks in the phosphorescence spectra and hence the zero point energies can be uniquely assigned. In these studies a particular dihalonaphthalene was dissolved in dilute quantities (usually less than 0.5 mol.%) as a guest in another dihalonaphthalene which served as the host matrix. Although the triplet energies of the guest are a sensitive function of the host crystal energies, the triplet state parameters just mentioned felI within a specified narrow range for a particular dihalonaphthalene. Thus the traps which were created in the chemicaIly mixed crystals could be assigned to a particular guest molecule. DBN was purchased from Pfaltz and Bauer. It was subjected to elution through a neutral alumina column, to recrystallization from ethanol and it was zone refined slowly at a rate of 2 cm h-l in a sealed tube for over 200 passes. Both DCN and BCN were synthesized from the respective 1-amino-ghalonaphthalene by the Sandmeyer reaction. The ether extract from the reacted diazotized mixture was washed, was decolorized and was dried over NazS04 and the ether was evaporated. Recrystallization of the resuIting solid residue from absolute ethanol gave approximately a 20% yield of the desired halonaphthalene. The material was then zone refined for an equiva-

162

lent of over 200 passes. The. middle sections of the material in the zone refining tubes were then sealed under vacuum and single crystals were grown from the melt by the standard Bridgman method. The crystals were carefully annealed at just below the melting temperature. The kinetic relaxation parameters of the total phosphorescence decay rate constants were obtained by several methods. The decay of the phosphorescence intensity measured subsequent to shuttering of the continuously applied excitation source was repetitively accumulated in a signal averager which was directly interfaced to a minicomputer equipped with a visual display. Using a linear regression program the resulting sum of the three decays was computer fitted to an equation describing the actual phosphorescence decay. The total phosphorescence decay rate constants from the individual sublevels of the triplet state obtained by the aforementioned shutter method were consistent with those obtained by the microwave-induced delayed phosphorescence technique first described in refs. 7 and 8 and with those obtained by the rapid passage method described by Winscom and Maki [ 91. The initial study involved the characterization of the triplet state parameters described earlier. Dilute concentrations of dihalonaphthalene were dissolved in host solvents which formed energetically deep traps. These included DCN, BCN and DBN in naphthalene-hs, naphthalene-ds and durene. The data presented in Table 1 show that the total phosphorescence rate constants and the zfs of the individual triplet sublevels are different for the three dihalonaphthalenes. Once the triplet state parameters were characterized, traps formed in neat single crystals of the dihalonaphthalenes could be examined and the results are presented in Table 2. The zfs and the kinetic behavior of the most prominent trap are given in the table. In the several traps observed for each neat single crystal, the individual traps have slightly different zfs and phosphorescence rate constants, although they are still very similar to those observed in the deep trap systems. DBN and DCN both yielded phosphorescence emission spectra having narrow linewidths (Fig. l), but the linewidths of the spectrum of BCN were much broader (Fig, 2). Presumably the reduced symmetry of BCN and the small difference in the size of the two substituted halogens would not allow the energetically preferred stacking orientation in the crystal formation. Rearrangements such as rotation about the in-plane long axis would not be easily allowed because of the large energy barrier, even though such a rotation would be the energetically preferred orientation. This was partiatly confirmed by allowing the BCN crystal to anneal over a period of several days at about IO” below the melting point of the crystal. Noticeable narrowing of the inhomogeneously broadened line was observed subsequent to annealing. In the neat DCN, the appearance of the band origin at 4943 A was very dependent on the past history of the crystal, particularly on impurity concentration and on the manner in which the crystal was grown (see Fig. 1). The lifetime of this band is much shorter than the lifetime of any of the

163 TABLE

1

Zfs, rate constants and the 0,O band of the phosphorescence , naphthalene-da and durene

of DCN, BCN and DBN in

naphthalene-ha

(MHz)

kl (s-l)

Guest

Host

0,O band (cm-‘)

Transitions

DCN

Naphthalene-hs

20202

3328 2485

19

16

0.71

Naphthalene-dg

20109

3370 2497 -

25

14

o-51

Durene

19877

3283 2491 -

72

21

0.62

19798

3276 -

41

22

O-46

Naphthalene-hs

20218

3157 2647

180

75

2.2

Durene

20222

3164 2670 -

160

60

2-o

Naphthalene-hs

20175’

2926 2829

267

164

11.1

Naphthalene-dg

20036

3182 2532 -

120

12

Durene

19908

270

120

BCN

DBN

k2 (s-l)

k3 (s-l)

O-60

8.4

aData taken from ref. 4. TABLE

2

Zfs, 0,O bands and rate constants of the first prominent localized state in neat single crystals of the dihalonaphthalenes

Dihalonaph DCN’

thalene

0,O band (cm-‘) (20231) 20198

‘ZYansitions (MHz) 3360

kl (s-l) (290) 16

k2 (s-l)

k3 (s-l)

(59) 7.5

-o ’ 75

2611 BCN

20214

3165 2675

170

110

6.9

DBNb

20164

2968 2827

251

180

10.2

aThe parameters in parentheses are assigned to the exciton origin. bData taken from ref. 4.

164

DCN

4.2

Ud

-4’

K

d-

> t

1.4 K

E !z

DBN

4.2

K

8

E

1.4 K

v

c j-J

8 if is

A

4.2 K

DCNlDBN

1.4 K I

4950

5050 WAVELENGTH

5150 th

Fig. 1. Phosphorescence spectra of neat single crystals of DCN and of DBN at 4.2 and 1.4 K. Spectra for chemically mixed DCN in a DBN single crystal at the two temperatures are also shown.

traps found either in neat DCN or in crystals in which DCN is the guest. We have tentatively assigned this peak to the exciton origin and are attempting to observe the exciton magnetic resonance at zero field. Although there are six possible guest-host combinations in the chemically mixed crystals, only DCN in DBN formed identifiable localized energy traps involving the guest molecule. This observation is peculiar considering that, in the neat crystals, the energies of the phosphorescence origin are BCN

4.2

K

1.4 K

-L& 4950

5050 WAVELENGTH

5150 tb,

Fig. 2. Phosphorescence spectra of neat BCN

at 4.2 and 1.4 K.

165

20 190, 20 214 and 20 231 cm-r for DBN, BCN and DCN respectively. The trend is consistent with arguments by Petruska [lo] in which the halogen substitutents would be predicted to cause a red shift due to first order inductive and second order conjugative and charge transfer effects. Table 3 gives the zfs, the trap depth and the kinetic parameters of the localized excitation of DCN in DBN crystals. The zfs and the rate constants of the origin at 20 164 cm-’ are consistent with the assignment of the origin as arising from a trap found in neat DBN (cf. Table 2 and Trap I in ref. 4). The next energetically lower origin, however, has zero field parameters characteristic of DCN (cf. Tables 1 and 2) and the trap intensity is dependent on the concentration of DCN in the DBN solvent. We have tentatively assigned this origin to phosphorescent DCN molecules. Attempts were made to observe BCN phosphorescence in DBN crystals. One of the phosphorescence origins found in neat DBN at 38 cm-l below the exciton band does vary somewhat in intensity with BCN concentration but a definitive assignment was not made. Because of the shallowness of the suspected BCN phosphorescence origin, the decay kinetics of the phosphorescence may be influenced by the thermal repopulation of the trap to the exciton band. A preliminary solubility study of the various solid solutions of dihalonaphthalenes showed that, at about 10 mol.% of DCN in DBN, the phosphorescence linewidths became dramatically broadened indicating limited solubility. This is consistent with results of Bellows and Prasad [6] in which the solubility was found to be limited to less than 10%. DCN in BCN showed broadened phosphorescence at a concentration of about 7 mol.%, whereas no noticeable line broadening was observed for BCN in DBN at these concentrations. Particularly in neat DCN and neat DBN, the presence of multiple trapping sites makes the characterization complex and only a portion of the ODMR results for one of several traps in each system is presented here. Following the crystallographic structural results in DCN, a complete examination of the dynamics of the trapping process using polarization measurements TABLE

3

Zfs, rate constants and trap depths for the phosphorescent single crystal of DBN at l-4 K 0,O origin (cm-‘)

Trap depth (from exciton

20164

28

20145

47

Trarnsitions

triplet states of DCN in a

(MHz)

kl (s-l)

k2 (s-l)

k3 (s-l)

290

90

25

origin) 5J 2900 -

3321

17

6.0

0*71

in ODMR, similar to those used by Dennis and Tinti [ ll] and by Lin and El-Sayed [ 121, will be presented, Also in progress is the use of pulse ODMR to study spin relaxation in the exciton trapping process. Acknowledgment We thank the National Institute of Health this research.

(GM 21770)

1 R. M. Hochstrasser, T. Li, H. N. Sung, J. E. Wessel and Chem.,

37 (1974)

85.

for supporting

A. H. Zewail,PureAppl.

2 R. Schmidberger and H. C, Wolf, Chem. Phys, Lett,, 16 (1972) 402. 3 R. M. Hochstrasser and A. H. Zewail, Chem. Phys., 4 (1974) 142. 4 A.M. Nishimura,A. H. Zewail and C. B.Harris,& Chem. Phys.. 63 (1975) 1919. 5 J. Trotter, Can. J. Chem., 39 (1961) 1574. 6 J. C. Bellows and P. N. Prasad, J. Chem. Phys., 67 (1977) 5802. 7 J. Schmidt, W, S. Veeman and J. H. van der Waals, Chem. Phys. Lett., 4 (1969) 341. 8 J. Schmidt, D. A. Antheunis and J. H. van der Wadis, Mol. Phys., 22 (1971) 1. 9 C. J. Winscom and A. H. Maki, Chem. Phys. Lett., 12 (1971) 264. 10 J. Petruska, J. C%em. Phys., 34 (1961) 1111;34 (1961) 1120. 11 L. W. Dennis and D. S. Tinti, Chem. Phys. Left., 23 (1973) 535. 12 C. T. Lin and M. A. El-Bayed. Chem. Phys., 4 (1974) 161.